Uncoupling of sustained MAMP receptor signaling
from early outputs in an Arabidopsis endoplasmic
reticulum glucosidase II allele
Xunli Lua, Nico Tintora, Tobias Mentzelb, Erich Kombrinka, Thomas Bollerb, Silke Robatzeka, Paul Schulze-Leferta,1,
and Yusuke Saijoa,1
aDepartment of Plant Microbe Interactions, Max Planck Institute for Plant Breeding Research, 50829 Cologne, Germany; and bBotanical Institute, University
of Basel, 4056 Basel, Switzerland
Edited by Frederick M. Ausubel, Harvard Medical School, Boston, MA, and approved October 22, 2009 (received for review July 15, 2009)
Recognition of microbe-associated molecular patterns (MAMPs), con-
served structures typical of a microbial class, triggers immune re-
sponses in eukaryotes. This is accompanied by a diverse set of
physiological responses that are thought to enhance defense activity
in plants. However, the extent and mechanisms by which MAMP-
induced events contribute to host immunity are poorly understood.
Here we reveal Arabidopsis priority in sweet life4 (psl4) and psl5
mutants that are insensitive to the bacterial elongation factor (EF)-Tu
epitope elf18 but responsive to flagellin epitope flg22. PSL4 and PSL5,
respectively, identify ␤- and ␣-subunits of endoplasmic reticulum-
resident glucosidase II, which is essential for stable accumulation and
quality control of the elf18 receptor EFR but not the flg22 receptor
FLS2. We notice that EFR signaling is partially and differentially
impaired without a significant decrease of the receptor steady-state
levels in 2 weakly dysfunctional gII␣ alleles, designated psl5-1 and
rsw3. Remarkably, rsw3 plants exhibit marked supersusceptibility
against a virulent bacterial phytopathogen despite nearly intact
coactivation of MAPKs, reactive oxygen species, ethylene biosynthe-
sis, and callose deposition in response to elf18, demonstrating that
these signaling outputs alone are insufficient to mount effective
immunity. However, rsw3 plants fail to maintain high transcript levels
of defense-promoting WRKY, PR1, and PR2 genes at late time points
(4 to 24 h) after elf18 elicitation. This points to an unexpected
separation between initial and sustained activation of EFR-mediated
signaling in the absence of proper glucosidase II-mediated endoplas-
mic reticulum quality control. Our findings strongly suggest the
importance of sustained MAMP receptor signaling as a key step in the
establishment of robust immunity.
EFR 兩 ER quality control 兩 LRR RLK 兩 plant immunity
A ctivation of pattern recognition receptors (PRRs) upon mi-
crobe-associated molecular pattern (MAMP) perception
leads to an enhanced state of immunity that limits invasion and
propagation of potential microbial intruders, termed MAMP-
triggered immunity (MTI) (1). MTI is associated with the activation
of a stereotypical set of cellular responses that occur from seconds/
minutes to hours/days upon elicitation. Early responses including
ion fluxes across the plasma membranes, reactive oxygen species
(ROS) spiking, and MAPK activation are generally followed by
ethylene production, transcriptional reprogramming, metabolomic
changes, and callose deposition (2–6). As recognition of different
MAMPs triggers largely similar host responses, it is presumed that
distinct PRR pathways converge on those signaling outputs. How-
ever, the physiological relevance of these MTI-associated events for
overall host defense activity and the mechanisms by which a single
receptor regulates signaling pathways leading to such diverse out-
puts remain elusive.
Because prolonged defense activation results in growth retarda-
tion, repression of abiotic stress responses, and/or cell death in
plants (5, 7–9), plants evolved specific mechanisms conferring
stringent control on abundance and activation/deactivation states of
immune receptors. As for intracellular immune receptors contain-
22522–22527 兩 PNAS 兩 December 29, 2009 兩 vol. 106 兩 no. 52
ing nucleotide binding and leucine-rich repeat (LRR) domains, a
widespread class of disease resistance proteins, the importance of
protein abundance control via cytosolic HSP90/SGT1/RAR1 chap-
erone complexes has been well documented (10). The requirement
of HSP90 and SGT1 has been demonstrated for nucleotide binding/
LRR protein functions in vertebrate innate immunity (11). How-
ever, it is still unclear whether these chaperones are engaged in
postactivation signaling and/or desensitization of the client immune
receptors.
In eukaryotic cells, folding and maturation of the majority of
membrane-localized proteins occur in the endoplasmic reticulum
(ER), where elaborate quality control ensures that only correctly
folded proteins are delivered to their functional sites. A major
branch of endoplasmic reticulum quality control (ERQC) relies on
Asn (N)-linked glycosylation on the client proteins (12–14). N-
glycosylation is catalyzed by the oligosaccharyltransferase (OST)
complex that transfers a preassembled glycan chain
(Glc3Man9GlcNAc2) to N residues in the sequon N-X-Ser/Thr (X ⫽
any amino acid except Pro) of acceptor proteins. Subsequent
trimming of terminal glucose residues by glucosidase I (GI) and
glucosidase II (GII) produces mono-glucosylated glycans
(Glc1Man9GlcNAc2) on the client proteins, thereby facilitating
their recognition and folding by the ER-resident chaperons caln-
exin (CNX) and calreticulin (CRT). Following this folding attempt,
GII-mediated removal of the outermost glucose residues releases
Man9GlcNAc2-conjugated client proteins from the chaperones.
Properly folded proteins are transferred to their functional sites,
whereas unfolded proteins are recognized by UDP-glucose:glyco-
protein glucosyltransferase (UGGT). UGGT attaches a glucose
residue to N-linked Man9GlcNAc2 glycans of client proteins, and
then facilitates the client proteins to enter reiterated rounds of
CNX/CRT-assisted folding (CNX/CRT cycle) (12–14). Severe loss
of OST, GI, or GII function causes lethality in plants as well as in
animals (12, 15–18). Arabidopsis plants carrying weak alleles of
these genes are viable, but show phenotypic alterations under
abiotic stress conditions (16, 18), suggesting a rate-limiting role of
the N-glycosylation pathway for the adaptation to these stresses.
In Arabidopsis, the LRR receptor-like kinases (RLKs) EFR
and FLS2, respectively, act as PRRs for the bacterial epitopes
elf18 and flg22 (19, 20). Here we present evidence that Arabi-
dopsis GII is required for stable accumulation and function of
EFR but not of FLS2. We further show that EFR signaling
outputs are partially and differentially impaired in weakly de-
Author contributions: X.L., P.S.-L., and Y.S. designed research; X.L., N.T., T.M., and Y.S.
performed research; X.L., N.T., E.K., S.R., and Y.S. contributed new reagents/analytic tools;
X.L., N.T., T.M., T.B., and Y.S. analyzed data; and X.L., P.S.-L., and Y.S. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1To whom correspondence may be addressed. E-mail: saijo@mpiz-koeln.mpg.de or
schlef@mpiz-koeln.mpg.de.
This article contains supporting information online at www.pnas.org/cgi/content/full/
0907711106/DCSupplemental.
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0907711106

Fig. 1. PSL4 and PSL5, respectively, identify GII ␤- and ␣-subunits that are
required for EFR but not FLS2 function. (A) WT (Col-0) and gII mutant seedlings
grown in the absence of sucrose (⫺ Suc) or presence of 100 mM sucrose (⫹ Suc)
without or with 1 ␮M flg22 (⫹ flg22) or elf18 (⫹ elf18). (B) Anthocyanin content
of seedlings grown as described in A, including the SD. (C) Schematic description
of the structure of GII ␤- and ␣-subunits (647 and 921 aa residues, respectively).
Positions of changes in the amino acid sequences in the mutant alleles are shown
(Bottom). (D) Immunoblot analysis of protein extracts from 2-week-old nonelic-
ited plants with the indicated antibodies. As for EFR, a long exposed blot is also
shown in the second panel. A Coomassie blue-stained blot is presented to verify
equal loading. Positions of molecular weight markers are shown (Right).
fective gII alleles, providing a genetic tool to dissect postrecog-
nition signaling events of EFR. In a previously isolated gII␣
allele, designated rsw3, EFR-mediated immunity to a bacterial
pathogen is compromised despite the coactivation of ROS,
MAPKs, ethylene biosynthesis, and callose deposition. However,
activation of defense gene expression is not maintained in the
mutant, pointing to an unexpected role of sustained activation of
PRR signaling for effective immunity.
Results
Identification of Arabidopsis priority in sweet life (psl) Mutants That
Are Insensitive to elf18 but Sensitive to flg22. Exposure to different
microbes or MAMPs leads to repression of flavonoid accumulation
in plants (8, 9), at the cost of the adaptation to various abiotic
stresses (21). For example, sucrose stress-induced anthocyanin
accumulation is blocked in the presence of flg22 or elf18 in
Arabidopsis seedlings (Fig. 1 A and B) (22). We have screened more
than 60,000 ethyl methanesulfonate-mutagenized M2 seedlings for
individuals that are defective in this negative crosstalk. Our screens
revealed ⬎50 psl mutants that show derepressed sucrose-induced
Lu et al.
anthocyanin accumulation in the presence of elf18 but retain
WT-like responsiveness to flg22 (Fig. 1 A and B). The results
indicate the existence of separate genetic requirements for the
function of the corresponding 2 PRRs despite their highly related
overall structure (1). The psl mutants are classified into at least 5
complementation groups, including novel efr alleles as well as
nonreceptor psl1, ps2, psl4, and psl5 mutants (Fig. 1 A and B) (22).
We described elsewhere that PSL1 and PSL2 identify CRT3
(AT1G08450) and UGGT (AT1G71220), respectively (22).
PSL4 and PSL5, Respectively, Define ␤- and ␣-Subunits of the ER-
Resident GII that Is Indispensable for Biogenesis of EFR but Not FLS2.
We identified PSL4 by positional cloning of the psl4–1 allele and
subsequent recovery of another independent allele, psl4–2 (Fig. 1C
and Fig. S1 A). PSL4 encodes the deduced ␤-subunit of the ER
lumen enzyme GII (At5g56360), the only homologue annotated in
the Arabidopsis genome (Fig. 1C). Consistent with the predicted
truncations in the large C-terminal portion of the protein, both psl4
mutants fail to accumulate the encoded protein (Fig. S1B). Besides,
psl4–1 plants transformed with a genomic GII␤ copy fully comple-
ment the psl phenotype (Fig. S1 D and E). Thus, we conclude that
GII␤ is required for EFR-mediated anthocyanin repression.
GII has been shown in animal cells to act as a heterodimer of
2 subunits, ␣ and ␤: GII␣ mediates the catalytic activity of the
PLANT BIOLOGY
enzyme whereas GII␤ directly interacts with and holds GII␣ in
the ER through its ER retention signal (23, 24). In view of the
functional interdependence between the 2 GII subunits, we
searched for gIIa alleles in the remaining psl mutants. A single
gene is annotated to encode a GII␣ homologue (At5g63840) in
Arabidopsis. Sequence analysis of the genomic locus revealed a
gII␣ allele, psl5–1, carrying a point substitution in the catalytic
domain (Fig. 1C). In addition, we tested MAMP responses in a
previously isolated gII␣ allele, rsw3, that results in another amino
acid substitution within the catalytic domain. The rsw3 allele has
been described to show a swollen root phenotype associated with
defects in cellulose biosynthesis at high temperature (30 °C) (16).
At the permissive temperature of approximately 22 °C, at which
rsw3 roots develop indistinguishably from WT, both gII␣ alleles
exhibit derepressed anthocyanin accumulation in the presence of
elf18 but not flg22 (Fig. 1 A and B). Both Ser residues substituted
in the gII␣ alleles are invariant in GII␣ orthologues not only from
other plants but also in mouse and Schizosaccharomyces pombe
(16). We verified cosegregation of the identified substitutions
and elf18 hyposensitivity in both gII␣ alleles (Tables S1 and S2).
Accumulation of the GII␤ subunit remains unaffected in these
gII␣ plants (Fig. S1B). Thus, although it is currently unknown
whether and how these substitutions influence GII catalytic
activity in planta, our genetic evidence indicates that both GII␣
and GIIß subunits are required for EFR but not FLS2 function.
In parallel, we have shown selective requirements of the ER-
resident CRT3, UGGT, and an OST subunit for stable accumula-
tion and thus function of EFR but not of FLS2 (22). Immunoblot
analysis of protein extracts derived from nonelicited plants revealed
that steady-state levels of EFR are greatly reduced without a
significant decrease in its mRNA levels in the presence of psl4
alleles (Fig. 1D and Fig. S1C). This is accompanied by a severe
defect in EFR-dependent binding capacity to the ligand elf26
(equivalent to elf18; Fig. S2). Thus, GII seems to promote stable
EFR accumulation at a posttranscriptional step during its biogen-
esis. Together, this is in accordance with the notion that GII acts in
concert with CRT3 and UGGT in an ER N-glycosylation pathway
that defines the biogenesis route of EFR. An apparently high-
molecular-weight form of EFR detected in psl4 plant lysates (Fig.
1D) might represent an under-trimmed N-glycan-conjugated
form(s) of the receptor, presumably as a consequence of low ER
GII activity in the mutants. Importantly, the abundance, apparent
size, and specific ligand binding of FLS2 remain unaffected in the
psl4 plants (Fig. 1D and Fig. S2), again pointing to the specific
PNAS 兩 December 29, 2009 兩 vol. 106 兩 no. 52 兩 22523
Fig. 2. MAMP-induced ROS spiking and MAPK activation in gII mutant
plants. (A and C) ROS spiking triggered in leaf discs of the WT, efr, fls2, and gII
plants at 100 nM elf18 (Left) or flg22 (Right), including the SD. (B and D) MAPK
activation in WT and gII seedlings upon application of water for 5 min (⫺) or
1 ␮M elf18 or flg22 for the indicated times. Positions of active MPK3 and MPK6
forms (Left) and molecular weight markers (Right) are indicated.
requirements of a GII-mediated ERQC step for EFR but not FLS2
biogenesis.
EFR-Mediated Signaling Is Severely Impaired in Strong psl4 (gII␤)
Mutant Plants. We next tested possible defects in characteristic
MAMP-induced responses in psl4 plants. The observed decrease in
EFR abundance, a complete size shift of the receptor (Fig. 1D), and
lack of GII␤ detection (Fig. S1B) demonstrate that these plants
carry strongly defective gII␤ alleles. Both psl4 plants fail to trigger
ROS spiking and MAPK activation in response to elf18 (Fig. 2 A
and B). These plants are also defective in PMR4/GSL5-dependent
callose deposition in the presence of elf18 (Fig. 3). In addition,
elf18-induced ethylene production is hardly detectable in psl4–1
plants (Fig. S3). However, both psl4 plants retain WT-like respon-
siveness to flg22 in all of the assays (Fig. 2 A and B, Fig. 3, and Fig.
S3), thus making it unlikely that the machineries directly executing
these responses are dysfunctional per se.
EFR-Mediated Signaling Is Partially and Differentially Impaired in
Weak psl5 (gII␣) Mutant Plants. We noticed that EFR accumulates
at WT-like levels in the absence or presence of elf18 elicitation in
both gII␣ alleles (Fig. 1D and Fig. S4A) that are defective in
elf18-dependent anthocyanin repression (Fig. 1 A and B). However,
EFR-dependent ligand binding activity is greatly and slightly re-
duced in psl5–1 and rsw3 plants, respectively (Fig. S4B). This
strongly suggests that dysfunction of GII␣ perturbs EFR function
per se, likely through improper folding of EFR. This is a unique
example in plants indicating that the ␣- and ␤-subunits of GII
indeed act in concert for an ERQC client protein. As one function
of GII␤ is to engage into the ER-retrieval mechanism the ␣-subunit
22524 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0907711106
Fig. 3. MAMP-induced callose deposition in gII mutant plants. Callose
deposits stained with aniline blue in the cotyledons of WT, efr, fls2, and gII
seedlings treated with water (⫺) or 2 ␮M elf18 or 1 ␮M flg22 for 24 h.
Representative photographs of more than 12 seedlings tested per sample are
shown.
that lacks both a transmembrane segment and a known ER
retention motif (Fig. 1C) (23, 24), it is conceivable that loss of GII␤
in psl4 plants (Fig. S1B) greatly decreases the ␣-subunit levels and
thus GII activity in the ER.
This raises the possibility that the residual ER GII activity in
the mutants is insufficient to maintain the amounts of recogni-
tion-competent receptor and thus full capacity to trigger down-
stream signaling. Elf18-induced ROS spiking is strongly reduced
below detectable levels in psl5–1 plants (Fig. 2C) even in the
presence of a high dosage (1 ␮M) of the ligand (Fig. S5).
Ethylene production is only slightly stimulated by elf18 (Fig. S3).
However, we observed a significant increase of MAPK activity
and callose deposits upon elf18 elicitation, albeit to a substan-
tially lesser degree compared with the WT plants (Fig. 2D, Fig.
3, and Fig. S6A). We note a possible discrepancy between our
results and the claim that MAPKs act upstream of ROS and
callose production in the FLS2 pathway (25). In contrast, all
FLS2-mediated signaling outputs tested are retained at the
WT-like levels (Fig. 2 C and D, Fig. 3, and Fig. S3).
Lu et al.

Fig. 4. Plants carrying gII mutations are highly susceptible to Pst DC3000. (A)
Growth of Pst DC3000 in 4-week-old plant leaves 3 d after spray infection with
bacteria at 1 ⫻ 109 cfu/mL. (B) MAMP-induced resistance of gII plants. Growth
of Pst DC3000 3 days after infiltration with bacteria at 1 ⫻ 105 cfu/mL on
4-week-old plant leaves pretreated with 1 ␮M elf18, flg22, or water (i.e.,
mock) for 24 h. SDs are given in A and B.
In rsw3 plants, we detected essentially intact activation of ROS
and MAPKs in response to elf18 as well as flg22 (Fig. 2 C and D),
although the duration of maximal elf18-induced MAPK activity
might be slightly shorter than WT plants (Fig. S6). We also
observed both elf18- and flg22-induced ethylene production and
callose deposits (Fig. 3 and Fig. S3). Ethylene measurement re-
vealed a WT-like elf18 dose dependence of rsw3 plants over a
1,000-fold tested dose range (1 ␮M to 1 nM; Fig. S3). These results
indicate that coactivation of these early and late EFR-signaling
outputs is insufficient for effective repression of anthocyanin ac-
cumulation (Fig. 1 A and B). Thus, these EFR-mediated signaling
outputs are uncoupled in the presence of the weakly defective gII␣
(rsw3) allele.
Plants Carrying Strong and Weak gII Mutations Are both Highly
Susceptible to a Virulent Bacterial Phytopathogen. To verify the
functional significance of the observed alterations in EFR-signaling
outputs, we next tested the immune activity of the gII plants.
Although previous studies have merely suspected a role of EFR in
plant immunity against the virulent bacterial phytopathogen
Pseudomonas syringae pv. tomato (Pst) DC3000 (6, 20), efr plants are
more susceptible than WT plants to this bacterium under our assay
conditions (Fig. 4A), in which we use a high dosage (109 cfu/mL) of
the bacteria for spray inoculation and keep the plants under high
humidity throughout the infection procedure (22) (see Materials
and Methods). Consistent with the extensive defects in the elf18-
induced responses examined, psl4 plants clearly show high suscep-
tibility upon challenge with Pst DC3000, at comparable levels to
Lu et al.
that of efr plants (Fig. 4A). Surprisingly, we detected a similar high
degree of bacterial multiplication in psl5–1 and rsw3 plants (Fig.
4A), although the latter plants retain elf18-induced ROS, MAPKs,
ethylene, and callose at WT-like levels (Fig. 2 C and D, Fig. 3, and
Fig. S3). We infer from this rsw3 plant phenotype that co-activation
of the tested EFR-mediated signaling outputs is insufficient to
mount effective defenses against the bacteria.
Elf18- but Not flg22-Induced Bacterial Resistance Is Compromised in
the Weakly Defective rsw3/gII␣ Allele. In view of a potentially wide
range of client membrane proteins that undergo GII-mediated
ERQC, it is conceivable that the sum of such indirect mutational
effects rather than specific defects in EFR-triggered immunity
renders rsw3 plants supersusceptible. To clarify this possibility, we
next directly assessed EFR-dependent defense activity in the mu-
tant plants. Plant pretreatment with elf18 or flg22 has been shown
to reduce the multiplication of Pst DC3000 (20, 26) and thus
provides a good indicator for MAMP-specific inducible defenses.
MAMP-induced resistance was defined as the decrease in bacterial
growth on elf18- and flg22-pretreated leaves as compared to
water-pretreated mock leaves (Fig. 4B). The elf18- but not flg22-
induced resistance is abolished in efr plants as well as in severely
EFR-depleted psl4–2 plants (Fig. 1D), verifying that elf18-induced
PLANT BIOLOGY
resistance occurs almost entirely through EFR (Fig. 4B). Our
analysis revealed that elf18-induced resistance is significantly im-
paired in rsw3 plants, albeit to a lesser extent compared with efr
plants, whereas fully functional flg22-induced resistance is retained
(Fig. 4B). This strongly suggests that EFR-triggered immunity is
selectively impaired in rsw3 plants.
EFR-Mediated Transcriptional Reprogramming Is Not Sustained in
rsw3 Plants. Our results predict that rsw3 plants are impaired at
another critical step in EFR-triggered immunity than the activation
of ROS, ethylene, MAPKs, and callose deposition that are gener-
ally considered as hallmarks for MTI (1). We therefore examined
potential alterations in elf18-induced defense gene expression in the
mutant, which would account for the immunocompromised phe-
notype (Fig. 4 A and B). Up-regulation of WRKY22 and WRKY29,
encoding members of the WRKY transcription factor family (27),
occurs downstream of MAPK activation in response to different
MAMPs (2, 25, 28, 29). We thus monitored expression of these
genes in WT and rsw3 seedlings during the time course of elf18
treatment. Both WRKY genes are rapidly induced within 1 h of elf18
application, and their transcript levels remain high over the 24-hour
time course analyzed in WT plants (Fig. 5). In contrast, rsw3 plants
fail to retain elevated WRKY transcripts after the rapid WT-like
increases within the first hour (Fig. 5). Together with the lack of a
detectable peak in strongly defective psl4–1 plants (Fig. S7), this
points to the existence of initial and secondary phases in EFR
signaling, of which the latter is affected in this weak gII␣ allele. The
initial activation of both WRKY genes is in good accordance with
the observed normal activation of the other early outputs, ROS,
ethylene, and MAPKs (Fig. 2 C and D and Fig. S3). Thus, the
observed transient nature of WRKY gene up-regulation suggests
that sustained activation, but not initial activation, of EFR signaling
is compromised in the mutant. Consistent with this, late activation
of PR1 and PR2, encoding defense-related proteins (30), is also
impaired in response to elf18. Conversely, flg22-mediated induction
of these genes is largely indistinguishable between WT and rsw3
plants (Fig. 5), again indicating a selective perturbation of EFR
function in the gII␣ allele.
Discussion
We show here that both the catalytic ␣-subunit and ER-retention
␤-subunit of GII are required for the biogenesis of functional EFR.
The identification of GII together with CRT3 and UGGT in our
present and parallel studies (22) has revealed several unique aspects
of ERQC in plants: (i) A branch of the ERQC system, namely the
PNAS 兩 December 29, 2009 兩 vol. 106 兩 no. 52 兩 22525

Fig. 5. Sustained transcriptional reprogramming on elf18 elicitation is
compromised in rsw3 plants. Quantitative RT-PCR analysis of 2-week-old
plants treated with 1 ␮M elf18 or flg22 for the indicated times. The relative
induction (in fold) is shown, with the gene/ACTIN values at 0 h in WT plants as
1. A representative data set is shown with SD of experimental replicates.
GII/CRT3/UGGT cycle, is selectively required for stable accumu-
lation of EFR but not of a structurally related PRR, FLS2. (ii)
Abundance control and quality control of a client receptor, EFR,
is uncoupled in weakly dysfunctional gII␣ and crt3 alleles. (iii)
EFR-mediated signaling outputs are partially and differentially,
rather than uniformly, impaired in these weak alleles. This latter
aspect provokes a previously unsuspected role of ERQC in the
modulation of postrecognition signaling of EFR.
Our genetic analysis has revealed important features of EFR-
mediated postrecognition signaling. In the presence of the psl5–1
(gII␣) allele, EFR-mediated ROS spiking is below detectable
levels despite substantial (albeit not full) activation of MAPKs
and callose deposition (Fig. 2 C and D and Fig. 3). Conversely,
in psl1–1 (crt3) plants, EFR-mediated ROS spiking occurs at
slightly lower levels than in WT plants despite nearly back-
ground-level activation of MAPKs and callose (22). These
results either contradict the notion deduced from FLS2-
mediated signaling that MAPKs act upstream of ROS spiking
(25) or suggest possible differences in the sequential order of
postrecognition signaling events initiated by EFR and FLS2.
More importantly, our data disfavor a simple threshold model in
which, e.g., more EFR signaling fluxes are required for callose
deposition than ROS spiking, as rejected in the psl5–1 plants
earlier. Rather, our data would favor a notion, at least in the case
of EFR, that these diverse signaling outputs are under the
control of separate signaling pathways. Lowered capacity of the
GII/CRT3/UGGT cycle would no longer allow EFR to govern
the whole signaling processes. A strong correlation is apparent
22526 兩 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0907711106
between the retained ligand binding capacity and retained
signaling outputs in those gII␣ and crt3 alleles (Fig. S4B) (22).
Considering the client receptor topology, it would be the LRR
domain that undergoes ERQC in the ER lumen and then is
exposed to the outer surface of the plasma membrane. The
folding quality of this presumed ligand binding domain of EFR
might determine its downstream signaling activity.
Substantial defects in EFR-triggered immunity (Fig. 4B) despite
essentially full co-activation of ROS, MAPKs, ethylene production,
and callose deposition in rsw3 plants (Fig. 2 C and D, Fig. 3, and Fig.
S3) hint at the existence of another as yet undefined key step in
MTI. It has been assumed that the aforementioned hallmark
outputs of PRR signaling collectively play a central role in MTI
signaling (1). In principle, our data do not disprove this, but propose
a mechanism beyond this concept. We note that EFR-triggered
immunity is significantly lowered, yet not abolished, in rsw3 plants
(Fig. 4B). This residual elf18-induced resistance is well correlated
with the observed coactivation of the 4 signaling outputs, thereby
merely disproving that co-activation of these outputs contributes to
MTI. Besides the anthocyanin derepression (Fig. 1 A and B), we
present evidence that sustained, but not initial, elf18-induced
up-regulation of WRKY22 and WRKY29 genes that encode defense-
promoting transcription factors (2) is impaired in the gII␣ plants
(Fig. 5). Furthermore, late induction of 2 genes encoding patho-
genesis-related proteins is also impaired in the mutant (Fig. 5).
These defects in EFR signaling would at least in part account for
the observed supersusceptibility to Pst.
Our findings highlight a separation in the presence of the weak
gII␣ allele between the initial induction phase and subsequent
sustained activation and/or signal amplification phase of EFR-
mediated signaling. However, the striking defects in the latter phase
become apparent in rsw3 plants between 1 and 4 h (Fig. 5), whereas
callose deposition is retained even 24 h after elf18 elicitation (Fig.
3). This again suggests the existence of parallel and/or multi-
branched signaling pathways emanating from EFR in WT plants, of
which a subset cannot be maintained in an active state in the
mutant. EFR produced in rsw3 plants might be unable to sustain
and/or strengthen defense signaling, possibly because of improper
subcellular partitioning, less ligand sensitivity, less stability of the
activated receptor per se and/or presumed receptor complex as-
semblies (31, 32), or combinations thereof. The importance of
sustained PRR activation for effective MTI is also supported by the
existence of bacterial effectors that directly suppress PRR function
(33, 34), given that initial PRR activation should precede the actions
of these effectors in the host cell. At present we cannot exclude that
the separation between the initial and sustained MTI activation
phases occurs at the level of another GII client protein(s) than EFR
that acts in its signaling pathway(s). However, such client(s) must be
dispensable for FLS2 signaling, given no detectable defects in
FLS2-triggered immunity (Fig. 4B) and FLS2 signaling outputs
(Fig. 1 A and B, Fig. 2 C and D, Fig. 3, and Fig. S3) in rsw3 plants.
Finally, our data point to the potential significance of sustained
MAMP receptor activation that maintains transcriptional repro-
gramming at a relatively late phase in mounting robust immunity.
This sustained activation and/or signal amplification phase might
involve the actions of MAMP-induced salicylic acid (35, 36) and/or
weak activation of resistance proteins (37).
Materials and Methods
Plant Materials and Growth Conditions. Arabidopsis M2 population used for psl
mutant screening is in the Col-0 glabrous1 (gl1) mutant background (Lehle
Seeds). efr-1 and fls2 mutants have been described previously (20, 26). The WT
control used was Col-0 unless otherwise stated. For the sucrose-MAMP crosstalk
assays, seedlings were grown under constant light in liquid medium containing
0.5⫻ MS for 3 d and then for a further 3 d with or without the addition of 100 mM
sucrose and MAMPs at the indicated concentrations. For MAPK/callose and gene
expression assays, seedlings were grown on 0.5⫻ MS agar plates or liquid me-
Lu et al.
dium, respectively, with 25 mM sucrose under 12 h light/12 h dark conditions for
10 to 14 d. Plants were grown on soil under 10 h light/14 h dark conditions for 4
to 5 weeks for ROS and bacterial inoculation assays.
Bioassays for MAMP-Induced Responses. Anthocyanin content in whole seed-
lings was determined as described (38), using at least 3 sets of more than 8
seedlings per treatment. ROS assays were performed essentially as described
previously (5) with the following modifications. Leaf discs (5 mm diameter)
excised from mature leaves were kept on water overnight before the lucif-
erase-based measurement of ROS generation triggered by the addition of
MAMPs at 100 nM unless otherwise stated. For MAPK and gene expression
assays, the whole seedlings were applied with elf18 or flg22 at 1 ␮M for the
indicated times. MAPK activation was detected by immunoblot analysis of
soluble proteins extracted from the seedlings in a lysis buffer described
previously (39) using anti-phospho p44/p42 MAPK antibody. For callose in-
duction, elf18 and flg22 were applied at 2 or 1 ␮M for 24 h, respectively.
Callose deposits were stained with aniline blue and visualized as described
(40). The MAMP concentrations used were optimized for consistent detection
of differences between WT and corresponding receptor mutant (efr or fls2)
plants: robust MAMP responses were hardly detectable without high varia-
tions in WT plants below the concentrations used under our conditions.
1. Boller T, Felix G (2009) A renaissance of elicitors: Perception of microbe-associated
molecular patterns and danger signals by pattern-recognition receptors. Annu Rev
Plant Biol 60:379 – 406.
2. Asai T, et al. (2002) MAP kinase signalling cascade in Arabidopsis innate immunity.
Nature 415:977–983.
3. Clay NK, Adio AM, Denoux C, Jander G, Ausubel FM (2009) Glucosinolate metabolites
required for an Arabidopsis innate immune response. Science 323:95–101.
4. Felix G, Duran JD, Volko S, Boller T (1999) Plants have a sensitive perception system for
the most conserved domain of bacterial flagellin. Plant J 18:265–276.
5. Gomez-Gomez L, Felix G, Boller T (1999) A single locus determines sensitivity to
bacterial flagellin in Arabidopsis thaliana. Plant J 18:277–284.
6. Kunze G, et al. (2004) The N terminus of bacterial elongation factor Tu elicits innate
immunity in Arabidopsis plants. Plant Cell 16:3496 –3507.
7. Asselbergh B, De Vleesschauwer D, Hofte M (2008) Global switches and fine-
tuning-ABA modulates plant pathogen defense. Mol Plant Microbe Interact 21:709 –
719.
8. Lozoya E, Block A, Lois R, Hahlbrock K, Scheel D (1991) Transcriptional repression of
light-induced flavonoid synthesis by elicitor treatment of cultured parsley cells. Plant
J 1:227–234.
9. McLusky SR, et al. (1999) Cell wall alterations and localized accumulation of feruloyl-
3⬘-methoxytyramine in onion epidermis at sites of attempted penetration by Botrytis
allii are associated with actin polarisation, peroxydase activity and suppression of
flavonoid biosynthesis. Plant J 17:523–534.
10. Shirasu K (2009) The HSP90-SGT1 chaperone complex for NLR immune sensors. Annu
Rev Plant Biol 60:139 –164.
11. Ye Z, Ting JP-Y (2008) NLR, the nucleotide-binding domain leucine-rich repeat con-
taining gene family. Curr Opin Immunol 20:3–9.
12. Anelli T, Sitia R (2008) Protein quality control in the early secretory pathway. EMBO J
27:315–327.
13. Hebert DN, Molinari M (2007) In and out of the ER: Protein folding, quality control,
degradation, and related human diseases. Physiol Rev 87:1377–1408.
14. Pattison RJ, Amtmann A (2009) N-glycan production in the endoplasmic reticulum of
plants. Trends Plants Sci 14:92–99.
15. Boisson M, et al. (2001) Arabidopsis glucosidase I mutants reveal a critical role of
N-glycan trimming in seed development. EMBO J 20:1010 –1019.
16. Burn JE, et al. (2002) The cellulose-deficient Arabidopsis mutant rsw3 is defective in a
gene encoding a putative glucosidase II, an enzyme processing N-glycans during ER
quality control. Plant J 32:949 –960.
17. Gillmor CS, Poindexter P, Lorieau J, Palcic MM, Somerville C (2002) Alpha-glucosidase
I is required for cellulose biosynthesis and morphogenesis in Arabidopsis. J Cell Biol
156:1003–1013.
18. Koiwa H, et al. (2003) The STT3a subunit isoform of the Arabidopsis oligosaccharyl-
transferase controls adaptive responses to salt/osmotic stress. Plant Cell 15:2273–
2284.
19. Gomez-Gomez L, Boller T (2000) FLS2: An LRR receptor-like kinase involved in the
perception of the bacterial elicitor flagellin in Arabidopsis. Mol Cell 5:1003–1011.
20. Zipfel C, et al. (2006) Perception of the bacterial PAMP EF-Tu by the receptor EFR
restricts Agrobacterium-mediated transformation. Cell 125:749 –760.
Lu et al.
Pathogen Inoculation and Growth Assays. Bacterial inoculation assays were
preformed as described previously (26) with the following modifications. Pst
DC3000 was sprayed onto leaf surface at 1 ⫻ 109 cfu/mL. Infected plants were
kept in a covered container for 3 d before harvesting leaves. A total of 12
surface-sterilized leaf discs (5 mm diameter) excised from 2 leaves of 6 plants
per genotype were randomly separated into 3 pools, and then subjected to the
quantification of leaf bacteria. For MAMP-induced resistance assays, plants
were syringe-infiltrated with 1 ␮M elf18, flg22, or water (i.e., mock) 24 h
before inoculation. Pst DC3000 suspension at 1 ⫻ 105 cfu/mL was syringe-
infiltrated into 2 to 3 leaves of 8 plants per genotype per treatment. Three days
after inoculation, a leaf disc (5 mm diameter) was excised from 12 represen-
tative leaves per treatment. These 12 leaf discs were separated into 3 pools and
then used to determine bacterial titers. These experiments have been re-
peated 3 times with the same conclusion.
ACKNOWLEDGMENTS. We thank Dr. Imre Somssich for useful comments on
the manuscript; and Dr. Steffen Rietz, Heidrun Häweker, Maret-Linda Kalda,
Brigitte Pickel, and Eva-Maria Reimer for technical assistance. The T-DNA
insertion line (psl4 –2) and rsw3 seeds were provided by the Arabidopsis
Biological Resource Center. This work was supported in part by the Max Planck
Society, grants from SFB670 (P.S.-L. and S.R.), and PhD fellowships from the
International Max Planck Research School Program (X.L. and N.T.).
21. Gould KS, Lister C (2006) Flavonoid functions in plants. Flavonoids: Chemistry, Bio-
chemistry, and Applications, eds Andersen ØM, Markham KR (CRC Press, Boca Raton,
FL), pp 397– 441.
PLANT BIOLOGY
22. Saijo Y, et al. (2009) Receptor quality control in the endoplasmic reticulum for plant
innate immunity. EMBO J 28:3439 –3449.
23. Trombetta ES, Fleming KG, Helenius A (2001) Quaternary and domain structure of
glycoprotein processing glucosidase II. Biochemistry 40:10717–10722.
24. Trombetta ES, Simons JF, Helenius A (1996) Endoplasmic reticulum glucosidase II is
composed of a catalytic subunit, conserved from yeast to mammals, and a tightly
bound noncatalytic HDEL-containing subunit. J Biol Chem 271:27509 –27516.
25. Zhang J, et al. (2007) A Pseudomonas syringae effector inactivates MAPKs to suppress
PAMP-induced immunity in plants. Cell Host Microbe 1:175–185.
26. Zipfel C, et al. (2004) Bacterial disease resistance in Arabidopsis through flagellin
perception. Nature 428:764 –767.
27. Shen QH, et al. (2007) Nuclear activity of MLA immune receptors links isolate-specific
and basal disease-resistance responses. Science 315:1098 –1103.
28. Libault M, Wan JR, Czechowski T, Udvardi M, Stacey G (2007) Identification of 118
Arabidopsis transcription factor and 30 ubiquitin-ligase genes responding to chitin, a
plant-defense elicitor. Mol Plant-Microbe Interact 20:900 –911.
29. Navarro L, et al. (2004) The transcriptional innate immune response to flg22. Interplay
and overlap with Avr gene-dependent defense responses and bacterial pathogenesis.
Plant Physiol 135:1113–1128.
30. van Loon LC, Rep M, Pieterse CMJ (2006) Significance of inducible defense-related
proteins in infected plants. Annu Rev Phytopathol 44:135–162.
31. Chinchilla D, et al. (2007) A flagellin-induced complex of the receptor FLS2 and BAK1
initiates plant defence. Nature 448:497–500.
32. Heese A, et al. (2007) The receptor-like kinase SERK3/BAK1 is a central regulator of
innate immunity in plants. Proc Natl Acad Sci USA 104:12217–12222.
33. Xiang T, et al. (2008) Pseudomonas syringae effector AvrPto blocks innate immunity by
targeting receptor kinases. Curr Biol 18:74 – 80.
34. Göhre V, et al. (2008) Plant pattern-recognition receptor FLS2 is directed for degra-
dation by the bacterial ubiquitin ligase AvrPtoB. Curr Biol 18:1824 –1832.
35. Mishina TE, Zeier J (2007) Pathogen-associated molecular pattern recognition rather
than development of tissue necrosis contributes to bacterial induction of systemic
acquired resistance in Arabidopsis. Plant J 50:500 –513.
36. Tsuda K, Sato M, Glazebrook J, Cohen JD, Katagiri F (2008) Interplay between MAMP-
triggered and SA-mediated defense responses. Plant J 53:763–775.
37. Kim MG, Geng X, Lee SY, Mackey D (2009) The Pseudomonas syringae type III effector
AvrRpm1 induces significant defenses by activating the Arabidopsis nucleotide-
binding leucine-rich repeat protein RPS2. Plant J 57:645– 653.
38. Teng S, Keurentjes J, Bentsink L, Koornneef M, Smeekens S (2005) Sucrose-specific
induction of anthocyanin biosynthesis in Arabidopsis requires the MYB75/PAP1 gene.
Plant Physiol 139:1840 –1852.
39. Saijo Y, et al. (2008) Arabidopsis COP1/SPA1 complex and FHY1/FHY3 associate with
distinct phosphorylated forms of phytochrome A in balancing light signaling. Mol Cell
31:607– 613.
40. Lipka V, et al. (2005) Pre- and postinvasion defenses both contribute to nonhost
resistance in Arabidopsis. Science 310:1180 –1183.
PNAS 兩 December 29, 2009 兩 vol. 106 兩 no. 52 兩 22527